490 BULLETIN 179, UNITED STATES NATIONAL MUSEUM 



landfall in the south from January 20 (Florida, Howell, 1932) to 

 29 (Louisiana, Oberholser, 1938). It is, therefore, one of the very 

 earliest of migrants. 



There is not, however, the marked regularity of appearance that 

 is characteristic of many other northward-bound species. Early 

 February would fit the Florida arrivals, on the average, better than 

 late January. For instance, at Melrose, Fla., the arrival date for 

 12 years averaged February 9 (Howell). 



In the Charleston, ^. C, area, Washington's Birthday is about the 

 time to expect this welcome summer resident, though the earliest record 

 is February 6, and Arthur T. Wayne (1910) has also noted them 

 on the 7th and 16th of that month. There are at least three dates 

 for the 16th. Strangely enough, they appear not to reach North 

 Carolina until "the latter part of March, while in most of the State, 

 the earliest dates are about the middle of April, with only an oc- 

 casional March record" (Pearson, Brimley, and Brindey, 1919). In 

 the vicinity of Kaleigh, N. C, it has been seen in March but once 

 in 25 years ! At Cape May, N. J., the arrival varies from March 27 

 to May 2, with the great majority of first dates occurring in April, 

 throughout the month (Stone, 1937). In New York and the Hudson 

 Valley, late April sees them arrive (Chapman, 1912), while the 

 species appears in Massachusetts and other parts of New England 

 from April 14 on (Forbush, 1929). Taverner (1934) gives no spe- 

 cific dates for arrival in Canada but mentions in another connection 

 that they arrive "early in spring" and are often caught by unseason- 

 able frosts and cold rains. The spring migration appears to be 

 about parallel in dates across the country; in Minnesota, for in- 

 stance, the birds sometimes arrive late in March but mostly early 

 in April (Eoberts, 1932). 



It will be seen, therefore, that the martin moves northward rather 

 leisurely for a range from late in January to early in May would 

 cover the extremes from Florida to the northern portions of the range. 

 April appears to be the time throughout much of the northern half 

 of the whole range across the country. As a general rule, the males 

 arrive in advance of the females and spend the interim in estab- 

 lishing themselves at old nesting boxes, feeding and preening, all 

 these activities being accompanied by a vociferous indulgence in vocal 

 effort. 



Nesting. — Before the advent of the white man the martin used 

 natural cavities of trees and cliffs for nesting sites. But even in 

 those distant days there was some bird-house nesting, for the Indians 

 were fond of these birds and, as Wilson (1831) says, "even the soli- 

 tary Indian seems to have a particular respect for this bird." He 

 gives an account of the methods used by the "Choctaws and Chicka- 



